Fabricio Alves Barbosa da Silva

BSP cost and scalability analysis for MapReduce operations

Data abundance poses the need for powerful and easy‐to‐use tools that support processing large amounts of data. MapReduce has been increasingly adopted for over a decade by many companies, and more recently, it has attracted the attention of an increasing number of researchers in several areas. One main advantage is that the complex details of parallel processing, such as complex network programming, task scheduling, data placement, and fault tolerance, are hidden in a conceptually simple framework. MapReduce is supported by mature software technologies for deployment in data centers such as Hadoop. As MapReduce becomes popular for high‐performance applications, many questions arise concerning its performance and efficiency. In this paper, we demonstrated formally lower bounds on the isoefficiency function for MapReduce applications, when these applications can be modeled as BSP jobs. We also demonstrate how communication and synchronization costs can be dominant for MapReduce computations and discuss the conditions under which such scalability limits are valid. To our knowledge, this is the first study that demonstrates scalability bounds for MapReduce applications. We also discuss how some MapReduce implementations such as Hadoop can mitigate such costs to approach linear, or near‐to‐linear speedups. Copyright

SparkBLAST: scalable BLAST processing using in-memory operations

BackgroundThe demand for processing ever increasing amounts of genomic data has raised new challenges for the implementation of highly scalable and efficient computational systems. In this paper we propose SparkBLAST, a parallelization of a sequence alignment application (BLAST) that employs cloud computing for the provisioning of computational resources and Apache Spark as the coordination framework. As a proof of concept, some radionuclide-resistant bacterial genomes were selected for similarity analysis.ResultsExperiments in Google and Microsoft Azure clouds demonstrated that SparkBLAST outperforms an equivalent system implemented on Hadoop in terms of speedup and execution times.ConclusionsThe superior performance of SparkBLAST is mainly due to the in-memory operations available through the Spark framework, consequently reducing the number of local I/O operations required for distributed BLAST processing.

Simulation step size analysis of a whole-cell computational model of bacteria

Understanding how complex phenotypes arise from individual molecules and their interactions is a major challenge in biology and, to meet this challenge, computational approaches are increasingly employed. As an example, a recent paper [1] proposed a whole-cell model Mycoplasma genitalium including all cell components and their interactions. 28 modules representing several cell functions were modeled independently, and then integrated into a single computational model. One assumption considered in the whole-cell model of M.Genitalium is that all 28 modules can be modeled independently given the 1 second step size used in simulations. This is a major assumption, since it simplifies the modeling of several cell functions and makes the modeling of the system as a whole feasible. In this paper we investigate the dependency of experimental results on that assumption. We have simulated the M.Genitalium cell cycle using several simulation time step sizes and compared the results to the ones obtained with the syst...

Identifying Drug Repositioning Targets using Text Mining

The current scenario of computational biology relies on the know-how of many technological areas, with focus on information, computing, and, particularly on the construction and use of existing Internet databases such as MEDLINE, PubMed and PDB. In recent years, these databases provide an environment to access, integrate and produce new knowledge by storing ever increasing volumes of genetic or protein data. The transformation and management of these data in a different way than from the one that were originally thought can be a challenge for research in biology. The problems appear by the lack of textual structure or appropriate markup tags. The main goal of this work is to explore the PubMed database, the main source of information about health sciences, from the National Library of Medicine. By means of this database of digital textual documents, we aim to develop a method capable of identifying protein terms that will serve as a substrate to laboratory practices for repositioning drugs. In this perspective, in this work we use text mining to extract terms related to protein names in the field of neglected diseases.

Vers une plate-forme MapReduce tolérant les fautes byzantines

Les pannes arbitraires sont inherentes aux calculs massivement paralleles tels que ceux vises par le modele MapReduce ; or les implementations courantes du MapReduce ne fournissent pas d’outils permettant de tolerer les fautes byzantines. Il est donc impossible de certifier l’exactitude des resultats obtenus au terme des traitements longs et couteux. Nous presentons dans cet article une architecture permettant de repliquer les tâches dans le modele MapReduce afin de garantir l’integrite des traitements et d’isoler les tâches defaillantes. Dans une premiere etude de performances nous avons evalue certains mecanismes lies a la replication. Une seconde etude, effectuee avec un prototype implementant l’ensemble de l’architecture, a permis de valider certains choix en montrant qu’il est possible de minimiser le surcout de la tolerance aux fautes byzantines.

Computational Modeling of Multidrug-Resistant Bacteria

Understanding how complex phenotypes arise from individual molecules and their interactions is a primary challenge in biology, and computational approaches have been increasingly employed to tackle this task. In this chapter, we describe current efforts by FIOCRUZ and partners to develop integrated computational models of multidrug-resistant bacteria. The bacterium chosen as the main focus of this effort is Pseudomonas aeruginosa, an opportunistic pathogen associated with a broad spectrum of infections in humans. Nowadays, P. aeruginosa is one of the main problems of healthcare-associated infections (HAI) in the world, because of its great capacity of survival in hospital environments and its intrinsic resistance to many antibiotics. Our overall research objective is to use integrated computational models to accurately predict a wide range of observable cellular behaviors of multidrug-resistant P. aeruginosa CCBH4851, which is a strain belonging to the clone ST277, endemic in Brazil. In this chapter, after a brief introduction to P. aeruginosa biology, we discuss the construction of metabolic and gene regulatory networks of P. aeruginosa CCBH 4851 from its genome. We also illustrate how these networks can be integrated into a single model, and we discuss methods for identifying potential therapeutic targets through integrated models.

The Challenge of Translating System Biology into Targeted Therapy of Cancer

Translational medicine has been leveraging new technologies and tools for data analysis to promote the development of new treatments. Integration of translational medicine with system biology allows the study of diseases from a holistic perspective. Cancer is a disease of cell regulation that affects genome integrity and ultimately disrupts cell homeostasis. The inter-patient heterogeneity is well characterized, and the scientific community has been seeking for more precise diagnoses in personalized medicine. The use of precision diagnosis would maximize therapeutic efficiency and minimize noxious collateral effects of treatments to patients. System biology addresses such challenge by its ability to identify key genes from dysregulated processes in malignant cells. Currently, the integration of science and technology makes possible to develop new methodologies to analyze a disease as a system. Consequently, a rational approach can be taken in the selection of the most promising treatment for a patient given the multidimensional nature of the cancer system. In this chapter, we describe this integrative journey from system biology investigation toward patient treatment, focusing on molecular diagnosis. We view tumors as unique evolving dynamical systems, and their evaluation at molecular level is important to determine the best treatment options for patients.

Modeling Gene Transcriptional Regulation: A Primer

The main goal of Systems Biology nowadays, from a broad perspective, is to explain how a living organism performs its basic activities of growth, maintenance, and reproduction. To attain this objective, investigation on a living phenomenon spans at least three levels of interactions: metabolic, transcriptional regulation, and signaling. A common aspect within these levels is biological phenomena control. In this text, we present an introduction to transcriptional regulation and its mathematical and computational modeling. From ubiquitous carbon source uptake to antibiotic resistance mechanisms exhibited by some bacteria, description of biological phenomena can always be associated with a certain control level, which is, directly or not, associated with transcriptional regulation. Our contribution here is to make explicit what are the consequences of making a transition from verbal (and visual) descriptive biological language to predictive domains of mathematical and computational modeling, showing what are the limitations and advantages this transition can imply.

A whole-cell simulator in GNU Octave

In 2012, a whole-cell simulator of bacteria Mycoplasma Genilalium was developed by a group of scientists in the University of Stanford. It was the first one who simulates the entire cycle of life. The simulator has 28 independent methods that are the main functions, like DNA Damage and Protein Folding. The communication among this methods is made by 16 variables named cellular states, like Metabolite and Geometry. From time to time, the methods update the cellular states and read them again. This time interval is called time step and the research group specified one second to it. This paper discusses the simulation porting from MATLAB to GNU Octave, a free software similar to the one used in the original simulator. The obtained partial results are presented.

The influence of the pentose’s pathway of the Clostridium Acetobutylicum on the production of butanol: Insights from mathematical modeling

This work is a pilot study for further analysis of the organism Trypanosoma cruzi (T. cruzi) and the influences of the Pentose’s Pathway on the parasite Clostridium acetobutylicum, already cataloged in the database of OptFlux program. We used the approach parcimonius Flux Balance Analysis (pFBA) to simulate the wild type organism and the mutant with an inhibition of the R_01056 reaction in pentose’s pathway. Results showed a reduction of approximately 1/3 of the biomass and 2/3 of the butanol production. This reduction shows the direct influence of the Pentose’s Pathway on the primary production of metabolites and the biomass generation from the Clostridium metabolites. This information prompted us to build in the future an SBML parameter file to represent the flow of T.cruzi pathways, which will be essential for the development of new drugs against.